On Thu, 19 Nov 2026 04:56:39 GMT, Jatin Bhateja <[email protected]> wrote:

>>> [@jatin-bhateja](https://github.com/jatin-bhateja) This is still filed as a 
>>> bug. But it does basically work as an extension, and as such should be an 
>>> **RFE**, right? Would it be ok to hold off on this and only integrate to 
>>> JDK29? What do you say [@PaulSandoz](https://github.com/PaulSandoz) ?
>> 
>> From my perspective, given the changes, the risk looks small (changing, 
>> accurately, to an RFE [*] does not change that for me). I think it comes 
>> down to all the other HotSpot fixes of higher priority that need to be 
>> reviewed and integrated before RDP1. So I am ok if you want to defer to 
>> after RDP1.
>> 
>> [*] it's an RFE that also fixes an issue, since it fixes an issue with 
>> operator reflection.
>
>> @jatin-bhateja This is still filed as a bug. But it does basically work as 
>> an extension, and as such should be an **RFE**, right? Would it be ok to 
>> hold off on this and only integrate to JDK29? What do you say @PaulSandoz ?
> 
> Hi @eme64 , its a bug which we decided to fix by extending the scope of 
> COMPRESS_BITS and EXPAND_BITS to sub-words.
> Changes looks low risk, as @PaulSandoz suggested and I second that opinion, 
> we can target this bug fix for jdk-28. Looking forward to your comments / 
> approval on compiler side of changes.

@jatin-bhateja I think you may be only partially reading what I said. I'll try 
and be clearer. I don't have as good an insight into the risk on the HotSpot 
side *and* there are also much higher priority issues in HotSpot that require 
reviewer attention before RDP1 (which I have more insight on). On more 
reflection, because of the latter (about priorities and reviewer time), let's 
defer and review/integrate soon after RDP1.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/33050#issuecomment-7018146372

Reply via email to